Labour market informality is widespread across developing countries. It is typically associated with lower wages, job instability, and lack of social protection. However, informal work arrangements also offer greater flexibility that can be desirable for workers with child-care responsibilities, especially in the presence of rigid labour regulations in the formal sector. This research project analyses the extent to which fertility decisions and the presence of children in the household affect the choices of women to work in the formal or informal sectors, and the ensuing life-cycle costs in terms of participation, accumulation of experience, hours, and earnings. Throughout this project, we use different sources of administrative data on fertility and labour market outcomes in Brazil to achieve the following objectives. First, we use these detailed individual-level data to estimate the career cost of children for women in Brazil. We emphasize formal labour market outcomes, but we also investigate the effects on informal employment outcomes. Second, we adopt a structural approach to analyse whether policies that seek to increase labour market flexibility are effective in preserving formal jobs following the birth of a child and improving labour outcomes over the lifecycle of women. In doing so, we hope to address key policy questions, such as to what extent labour market rigidities increase the career cost of children for women relative to men. Moreover, we aim to investigate which policy instruments are more effective in reducing these costs, in particular policies that introduce greater flexibility vis-à-vis policies that extend maternity leave or provide greater availability of childcare.
In the first half of the project, we have put together comprehensive, never-used information on more than 60 million births in Brazil, linked to an identified data set of parents, which we have then merged to two additional administrative data sets: (i) RAIS, a matched employer-employee dataset that contains the universe of formal labour contracts, which provides detailed, long-term longitudinal histories of employment in the formal sector of the labour market; and CadUnico, which is a unified registry of federal social welfare programs in Brazil. This has been a very time intensive endeavour, but which provided us with an extremely rich data to answer our research questions. After concluding this data construction stage, we conducted careful analysis of the impacts on mothers and fathers from the birth of their first child following the most recent literature on difference-in-differences design to fully explore the richness of the data. Subsequently, we have extended the reduced-form analysis by performing different heterogeneity analyses, as well as adapting our framework to cross-sectional data. In the second half of our project, we have developed and estimated a structural model of fertility and labour supply decisions across the formal and informal sectors of the labour market.
Our reduced-form empirical results show a substantial child penalty in the formal sector. After birth, mothers’ outcomes (presence in the formal sector, months formally employed and formal wages) decline up to 40% relative to the matched control group of women with no births. These gaps narrow down relatively fast compared to richer countries, such as Denmark, but a gap of around 20% remains even after 2 or 3 years. Interestingly, these costs tend to be stronger for women in municipalities with higher informality, but the convergence is faster than in municipalities with low informality. In what follows, we highlight some of the results from our structural approach. Our parameter estimates show that full-time work imposes substantial utility costs arising from the presence of young children in the household. Considering the overwhelming concentration of full-time jobs in the formal sector, this finding demonstrates the role that informality plays in providing labour supply flexibility to mothers. By incorporating labour market frictions in our framework, we are also able to show the extent to which demand for part-time work in the formal sector is limited. Our estimates suggest that approximately 99% of job offers in the formal sector are full-time. In addition, our counterfactual exercises demonstrate how children are a relevant constraint in life-cycle labour supply of women not only in terms of participation and full/part-time work. In fact, our results suggest that fertility is responsible for up to a 2-percentage point reduction in the rate of work in the formal sector, and for an increase in the rate of part-time work across all employment categories, but more strongly so in informal and self-employment.
